This article proposes a novel way of measuring cross-national changes over time in the outputs of social security systems. Traditional approaches to the comparative analysis of social security systems use expenditure levels, regime types or poverty and inequality rates to rank countries and map change over time. All these approaches encounter the problem of determining how much of the observed change is due to internal developments within the social security system, and how much due to exogenous social and economic factors. Taking the example of public pensions in five European countries since 1950, this article demonstrates how formal social security rules can be used in a simulation model to evaluate changes in public pension payments for a variety of hypothetical individuals characterised by different levels of lifetime income. This procedure produces direct measures of the impact of changes in social security systems which are entirely independent of exogenous developments in social and economic structures. This new method reveals the ‘pure’ effect of internal social security system development over time.